© 2006 Nature Publishing Group http://www.nature.com/natureimmunology
Maintenance and modulation of
T cell polarity
Matthew F Krummel 1 & Ian Macara 2
As T cells move through the lymphatics and tissues, chemokine receptors, adhesion molecules, costimulatory molecules and
antigen receptors engage their ligands in the microenvironment and contribute to establishing and maintaining cell polarity.
Cytoskeletal assemblies, surface proteins and vesicle traffic are essential components of polarity and probably stabilize the
activity of lymphocytes that must negotiate their ‘noisy’ environment. An additional component of polarity is a family of polarity
proteins in T cells that includes Dlg, Scrib and Lgl, as well as a complex of partitioning-defective proteins. Ultimately, the
strength of a T cell response may rely on correct T cell polarization. Therefore, loss of polarity regulators or guidance cues may
interfere with T cell activation.
T cells traffic through a signal-rich environment. In the blood, they receive
cues from chemokines, which permit attachment and extravasation into
tissues. In secondary lymphoid environments, receptors on the T cell surface
integrate signals from ligands bound to other cells or to the extracellular
matrix and from gradients of soluble mediators. At the sites of effector
function, T cells encounter secreted products of activated macrophages,
neutrophils and natural killer cells, which may result in polarized or nonpolarized
stimulation of cytokine receptors on T cells. Underlying a successful
T cell response, then, is the ability to ‘prioritize’ these cues and subsequently
orient effector mechanisms toward the optimal targets. This review discusses
polarity ‘themes’ in T cells and their potential inter-relationships
and will postulate how polarity is modulated when T cell receptor (TCR)
stimuli are encountered.
Polarity ‘themes’ in motile lymphocytes
Like a musical partita, the five systems of T cell morphology are variations
on a shared polarity ‘theme’. The actin cytoskeleton, tubulin cytoskeleton,
surface proteins, vesicle traffic and conserved polarity proteins are all polarized
relative to the motility axis. Among the best studied of those systems are
the actomyosin cytoskeleton and the GTPases associated with promoting
actin nucleation (Fig. 1a). Actin fibers are inherently polar fibers with their
barbed (‘plus’) end at the site of monomer addition and in the direction of
actin-based protrusion. That constitutes the leading edge in lymphocytes
and is probably very similar to the lammellipodial regions generated in
other motile cell types, in that free actin filaments drive protrusion independently
of myosin II crosslinking activities 1 . However, behind the leading
edge and extending through the midbody is a zone of high myosin II
1 Department of Pathology, University of California at San Francisco, San
Francisco, California 94143-0511, USA. 2 Center for Cell Signaling, Department
of Microbiology, University of Virginia School of Medicine, Charlottesville,
Virginia 22908-0577, USA. Correspondence should be addressed to M.F.K.
(email@example.com) or I.M. (firstname.lastname@example.org).
Received 7 August; accepted 14 September; published online 19 October
accumulation 2 in which it is likely that actin is organized by myosin in
rods that are mostly parallel. In epithelial cells, this region corresponds to
the lamella, and the myosin II in this zone causes retrograde flow of actin
filaments and attached membranes toward the rear of the cell 1 . Rearward
movement of surface-bound beads toward the uropod has demonstrated
that retrograde cortical flow also occurs in T cells 3 . Myosin II ‘motors’ are
probably responsible for retrograde flow, as shown by the accumulation of
myosin II at sites of leading edge retraction and the subsequent retrograde
flow of those motors into the uropod 2 .
The microtubule cytoskeleton and related kinesin and dynein motors
comprise a second polarized cytoskeleton (Fig. 1b). Like actin filaments,
microtubules are inherently polarized. Microtubules are organized around
a microtubule-organizing center (MTOC). That arrangement creates
another level of polarity, as the MTOC is invariably in the uropod and
behind the nucleus.
Surface receptors and perhaps membrane lipid rafts are also polarized
such that some are localized toward the uropod, whereas others, activated
integrins in particular, are likely to cycle toward and accumulate at the leading
edge 4 (Fig. 1c). Actomyosin and membrane movement from the front
of the cell to the rear of the cell probably accounts for the general accumulation
of many surface receptors at the uropod: TCRs, CD2, CD43 and CD44
accumulate at the uropod before activation 5–8 ; ICAM, CD43 and CD44
in particular are probably localized as a result of their association with
the C terminus–phosphorylated form of ezrin-radixin-moesin (cpERM),
which associates with actin 5,6 . In comparison, adhesive receptors such as
LFA-1 are selectively retained along the ‘midbody’ of the cell by adhesion 4 .
Cholesterol-rich rafts may also be polarized along with those receptors 9 .
Integrins may also be specifically recycled to the leading edge by as-yetunidentified
selective vesicular traffic from the uropod to the leading edge
(Fig. 1d). Cycling is important, as forward motility requires the internalization
of highly adhesive integrins at the front and midbody at the uropod
and that they be recycled ‘forward’ to contribute to new projections. It is
highly probable that endocytosis and exocytosis are generally polarized
along the axis of movement and also reinforce other systems of polarity
by asymmetrically depositing material at one end of the cell or the other.
NATURE IMMUNOLOGY VOLUME 7 NUMBER 11 NOVEMBER 2006 1143
a b c d e
© 2006 Nature Publishing Group http://www.nature.com/natureimmunology
and dissembly retrograde
Much vesicle movement probably involves the Rab family of proteins that
direct the fusion of vesicles with the plasma membrane and also associate
with specific myosins or kinesins for directional transport. The targeting
and docking of vesicles with membranes involves the recognition of vesicle
soluble N-ethylmaleimide-sensitive factor attachment protein receptors (v-
SNAREs) by target SNAREs (t-SNAREs) that are enriched at specific sites
along the membrane. SNAREs typically consist of syntaxin and SNAP proteins,
some of which have been identified in T lymphocytes 10,11 . Polarized
distribution of t-SNAREs at the leading edge of motile T cells remains to
be assessed, but would be predicted to ‘encourage’ delivery of vesicles in
the direction of motility.
The proteins scribbled (Scrib), lethal giant larvae (Lgl) and discs
large (Dlg), together with partitioning-defective (PAR) polarity proteins
(Fig. 1e), have emerged as a fifth set of participants regulating many aspects
of lymphocyte biology. Scrib, Lgl and Dlg proteins were first identified by
genetic screens in flies or worms and are key conserved elements required
for polarization of cells in many contexts. For example, Scrib was identified
in a screen for genes involved in apical versus basal polarity in drosophila
embryonic epithelium, but it also functions in synaptic vesicle dynamics 12
and acts as a tumor suppressor 13 . Scrib interacts genetically with the two
other tumor suppressors Lgl and Dlg, which are also essential for maintaining
apical versus basal polarity. Somewhat unexpectedly, no existing
evidence indicates the involvement of mammalian Scrib protein in the
maintenance of apical versus basal polarity, but Scrib has been linked to
the regulation of planar polarity of the inner ear epithelium 14 , in epithelial
cell-cell adhesion 15 , in endocytosis and recycling of G protein–coupled
receptors (GPCRs) 16 and, most unexpectedly, in T cell polarization 17 .
PAR proteins were first identified in a screen for genes affecting the first
asymmetric division of the Caenorhabditis elegans zygote. Of the six PAR
genes that have been cloned, those encoding PAR-3, PAR-6 and atypical
protein kinase C (aPKC) interact both genetically and physically. PAR-6
seems to function at least in part as a targeting subunit for aPKC, to which it
binds constitutively. The PAR proteins also interact with Scrib, Lgl and Dlg
proteins. For example, PAR-6 can recruit Lgl, which is then phosphorylated
by aPKC 18–20 .
Polarized assemblies in motile lymphocytes
T cells have a directional axis of motility, and a fundamental driving force
for the polarization and migration of T cells is the dynamic remodeling
of actin cytoskeletal elements. In most cell types that have been studied,
remodeling is controlled by the Rho family of GTPases through a plethora
of effector proteins. Rac and Cdc42 GTPases, in particular, have been associated
with the protrusion of the leading edge and in the orientation of
Figure 1 Five polarity systems in motile T lymphocytes. (a,b) Polarity can be described from the perspective of the actomyosin cytoskeleton (a) or tubulin
cytoskeleton (b), which are polarized relative to the direction of migration (green arrow). (c) A front-to-back ‘footprint’ of integrins on substrates (red shading)
and uropodal accumulation of receptors further delineates the polarity. (d) Both directional migration and the associated cycling of integrins require the
movement of material through the cytoplasm by means of an endocytic-exocytic loop. (e) Evidence places evolutionarily conserved polarity proteins in distinct
locales during migration.
migration (Fig. 2a). Overexpression of mutant forms of either GTPase
results in loss of polarity and defective chemotaxis toward the chemokine
SDF-1 (refs. 21,22). ‘Downstream’ effectors of Rac and/or Cdc42 include
the Wiskott-Aldrich syndrome protein (WASP), neural WASP and the
related WAVE (also called Scar) family of proteins, which induce actin
nucleation through the complex of Arp2 and Arp3. WASP is required for
optimal chemotaxis 23,24 , but evidence has indicated WAVE2 is a predominant
participant in hematopoietic cell migration 25–27 . WAVE2 is required
for actin cap formation at the immunological synapse (IS) and forms a
complex with the Hem-1 linker protein 26 . Hem-1 is required for actin
protrusions in neutrophils and augments leading-edge identity by inhibiting
activation of the myosin light chain at that site 25 .
In contrast to Rac and Cdc42, which promote actin nucleation, the
GTPase RhoA activates a protein kinase, ROCK, that phosphorylates
myosin light chains and thus increases the contraction of actin filaments
relative to each other (Fig. 2b). Rho is ‘downstream’ of chemokine GPCRs
and is required for increases in integrin affinity 28,29 . Thus, it is likely that
in contrast to Rac and Cdc42, Rho proteins located at the site of GPCR
signaling activate myosin, which triggers sliding of this membrane patch
toward the cell uropod, thereby generating forward thrust in the rest of the
cell. Although that model is likely, the system is more complex locally, as
a Rac complex with its effector PAK have also been linked to the chemotaxis
of both macrophages and T cells 30,31 . As Rac often antagonizes Rho
function 32 , the details of how these GTPases function in concert remains
In T cells, Rho is also required for uropod formation, possibly because
of its involvement in augmenting phosphorylation of ERM 33 . That function
is important, because cpERM in the uropod may tether a complex
that aids in recycling components back to the leading edge through vesicle
trafficking mediated by ADP-ribosylation factor (Arf; Fig. 2c). Arf family
GTPases control vesicle movement during motility in autologous systems 34 ,
although the function of these proteins in T cells has not been established.
The PDZ domains of mammalian Scrib associate with the C terminus of
the Rac guanine-exchange factor βPIX 35 , which in turn binds to GIT1,
a scaffold protein with an Arf GTPase-activating protein domain. This
Scrib-βPIX-GIT1 complex has been linked to the recycling of GPCRs 16 in
autologous systems, suggesting a similar function in T cells. Scrib is localized
in the uropod 17 , but how does that localization occur? In drosophila
neurons, Dlg determines the localization of Scrib, and these two proteins
form a complex that is mediated via the scaffold protein ‘GUK-holder’ 36 .
A mammalian homolog of GUK-holder has been identified, but it has
not yet been shown to be functionally equivalent to the fly protein, and
its expression in T cells is untested. However, human Dlg is targeted to
1144 VOLUME 7 NUMBER 11 NOVEMBER 2006 NATURE IMMUNOLOGY
© 2006 Nature Publishing Group http://www.nature.com/natureimmunology
membranes by its association with ERM 37 , and Dlg proteins are found in
the T cell uropod alongside cpERM proteins 17 . Notably, Scrib-deficient
T cells fail to maintain polarity and cease movement, consistent with requisite
involvement of these proteins in a motility cycle 17 .
A final functional cassette is probably in the motile cell and consists of
PAR-3, Lgl, PAR-6 and aPKC-ζ which are all localized in the midbody 17 .
In epithelial cells, Lgl can associate not only with myosin II and PAR-6
but also with syntaxins, which control vesicle delivery to specific target
membranes. As mentioned above, PAR-6 seems to function at least in part
as a targeting subunit for aPKC, to which it binds constitutively. Thus, in
drosophila embryonic epithelial cells, PAR-6 can recruit Lgl, which is then
phosphorylated by aPKC 18–20 , which in turn triggers disassociation of the
complex from the cell cortex. The PAR-6–aPKC complex can also bind to
and phosphorylate the ubiquitin E3 ligase Smurf1, one target of which
is RhoA 38 . A key function of the PAR-6 polarity cassette might therefore
be to trigger local degradation of RhoA, which would reduce actin contractility
and block Rho-induced phosphorylation of cpERM proteins. In
addition, PAR-6–aPKC binds to and phosphorylates PAR-3 (refs. 39–41).
As PAR-3 is capable of binding to and spatially restricting the activity of
the Rac guanine-exchange factor Tiam1, the PAR-6–aPKC–PAR-3 interaction
probably also regulates actin dynamics 42,43 . Tiam1 was first identified
as a T cell lymphoma invasion and metastasis gene 44 . The association of
PAR-3 with LIM kinase (LIMK) may further regulate actin in the area
through LIMK regulation of cofilin 43 . Variations in the composition of
PAR-6–aPKC complexes may create unique actin dynamics in the midbody
that facilitate both forward motility through actomyosin regulation and
vesicle trafficking through syntaxin regulation (Fig. 2d). The nature of
those complexes remain to be elucidated.
Cytoskeletal polarity in the IS
In a signal-rich environment, interrelated feedback systems like those
described above are important for stabilizing cellular activity. In contrast,
TCR signaling initiates a transformation from high motility to the stable
IS 45,46 . How might that signal create the conditions in which proteins are
repolarized, and how is a stable IS maintained?
The reorientation of the T cell during engagement with an antigen-presenting
cell (APC) and formation of the IS are reminiscent of the switch
in polarization that occurs during the epithelialization of mesenchymal
cells. In both cases, cells undergo substantial morphological changes, which
are driven by remodeling of the actin and microtubule cytoskeletons and
by relocalization of polarity proteins. During the mesenchymal-epithelial
transition, anterior-posterior polarity is converted into apical-basal
polarity. Microtubules and actin reorganize to form bands around the cell
periphery, the Rac GTPase redistributes from the leading edge to the lateral
membrane, and polarity protein such as PAR-3, Scrib, and Dlg translocate
to the cell-cell junctions. A functionally similar switch occurs in T cells:
the uropod at the posterior of the T cell disappears 46 and polarity proteins
such as Scrib and Dlg transiently redistribute from the uropod to the front
of the cell where the IS forms 17,47,48 .
A first step in this transformation may be the disassembly of existing
cytoskeletal structures (Fig. 3a). TCR signaling transiently leads to the
dephosphorylation of cpERM, and that modification results in cytoskeletal
relaxation, probably because of release of the membrane-actin linkage
anchored by ERM 49 . That occurs through Rac activation 49 , which may
antagonize Rho-induced ERM phosphorylation 33 . Such relaxation may
release cell surface receptors and proteins such as Dlg and Scrib, which can
then traffic to the leading edge. TCR signaling also induces the calciumdependent
phosphorylation of myosin II (ref. 2). Myosin heavy-chain
phosphorylation results in loss of myosin filamentation 50 , probably releasing
the aligned actin filaments in the midbody and uropod and permitting
reorganization of the components formerly located in those regions.
At the leading edge, TCR signaling induces actin complexes and the early
assembly of microclusters of TCRs 7,51 . Microclusters probably arise as a
result of TCR-induced activation of the Lat and SLP-76 adaptor proteins,
which trigger WASP and PAK activation, which promotes actin assembly
52–54 (Fig. 3b). However, Dlg, which is found at the center of the contact,
a b c d
Adherent contractile zone Uropodal recycling
Regulated midbody cytoskeleton
Syntaxin Myosin II
Figure 2 Distinct zones in motile lymphocytes. Various functional compartments serve to stabilize crawling activity. All these molecules have been found
in the locations here except those in gray italics, which have not yet been assessed in T cells. (a) A protrusive complex that uses a WASP family member to
activate actin elongation is probably associated with the Hem-1 molecule in a cassette, which inhibits myosin light-chain activation. N-WASP, neural WASP.
(b) A contractile cassette ‘downstream’ of chemokine receptors and integrins may use Rho to activate ROCK and thereby induce myosin II contractility.
Additional signaling (such as via phosphatidylinositol-3-OH kinase) is not presented here. MLC, myosin light chain. (c) The uropod is proposed to be the site
of a cpERM-based internalization ‘engine’. The cpERM-mediated recruitment of Dlg may also recruit Scrib via a GUK-holder (not yet identified in T cells).
Scrib localization in the uropod in turn may recruit and activate Arf proteins via the assembly of βPIX-GIT complexes. (d) PAR-6–aPKC–based cassettes
throughout the cytoplasm may prevent actin elongation through PAR-3-mediated binding of LIMK or Tiam1. Alternatively, PAR-6–aPKC complexes can
interact with Lgl, which interacts with syntaxins and myosin.
NATURE IMMUNOLOGY VOLUME 7 NUMBER 11 NOVEMBER 2006 1145
© 2006 Nature Publishing Group http://www.nature.com/natureimmunology
Arrest and cytoskeletal relaxation
Assembly of microclusters and polarity modulators
Crumbs– ? Pals1– PAR-6–aPKC
in the central-supramolecular activation cluster (c-SMAC), during the first
minutes of contact 17,47,48,55 , is probably another participant in this dynamic
process. Dlg associates directly with the Src homology 3 domains of both
the Zap70 and Lck tyrosine kinases 48,55 . In that way, Dlg probably senses
the proximity between Lck and Zap70 that occurs during clustering of
TCR and CD4 proteins and might even promote further aggregation of
those proteins (Fig. 3b). Dlg also interacts with microtubule kinesins 56 and
thus may act to initiate repolarization of the microtubule-based cytoskeleton.
Finally, the Lck-Zap70-Dlg complex also seems to include WASP 48 ,
and formation of that complex may help to move WASP into proximity
of active Cdc42. Transient phosphorylation of Zap70 and Lck when the
TCR is activated may trigger conformational changes in that complex and
promote its alteration from a motile to an IS form. Loss of Dlg from the
IS at later times 17 is consistent with loss of active Lck in the late IS 57 . It is
perhaps notable that elimination of Dlg by means of short hairpin RNA in
cell lines can alternatively augment 47 or attenuate 48 T cell activation. Those
observations are perhaps indicative of the importance of polarity in making
T cells receptive to signals in the first place or, alternatively, in stabilizing the
newly emerging signal, and those disparate results may reflect the polarized
nature of the stimulation provided in the various experiments.
Like Dlg, Scrib moves from the uropod, is transiently localized in the
nascent IS and subsequently moves to a position mainly outside the IS 17 . In
P MyoII (hexameric)
WASP TCR-actin complexes
Figure 3 Early events in polarity modulation. Specific multiprotein
complexes are suggested to mediate changes in polarity and morphology.
(a) Cytoskeletal relaxation is thought to occur as a result of cpERM
dephosphorylation as well as debundling of myosin II (MyoII) filaments.
(b) Specific multiprotein complexes then assemble at the T cell–APC
membrane contact site. These include TCR-actin complexes, Scrib
complexes (perhaps containing βPIX and PAK), TCR-Dlg complexes (perhaps
nucleating actin) and crumbs complexes (perhaps recruiting Pals and PAR-6
complexes). Molecules in gray italics have not yet been assessed in T cells,
but all others have been localized as presented here. All interactions are
not always presented here (for example, Zap70 has been linked to WASP
through multiple pathways). (c) Active transport is suggested to complete
repolarization. Myosin motors may be instrumental in recruiting uropodal
components along the membrane or through the cell. Similarly, microtubulebased
motors may recruit molecules toward or within the IS. PI(3)K,
the IS, Scrib may function to recruit the Rac and Cdc42 guanine-exchange
factor βPIX (Fig. 3b). That protein has been shown to recruit PAK to the
IS in T cells 58 . Therefore, defects in Scrib-deficient T cells 17 and in T cells
expressing dominant negative PAK 59 may reflect a requirement for a ScribβPIX-PAK
complex during T cell activation.
Finally, the regulation of PAR-6-based cassettes may represent a more
complex mechanism of influencing local polarity in response to signaling.
The PAR-6–aPKC heterodimer can bind several alternative partners
through its PDZ domain, including Lgl, PAR-3 and Pals1 (protein associated
with Lin7) 60 . In eukaryotes, Pals1 is recruited to the cell surface by
crumbs, the mammalian homolog of Lin7, and functions to recruit the
PAR-6–aPKC complex to the cell cortex. In the case of T cell IS, crumbs1
is constitutively localized in the IS 17 and thus may lend ‘apical identity’ to
this surface (Fig. 3b). The mechanism by which crumbs is localized to the
IS, however, is yet to be determined.
Release of cytoskeletal tension (Fig. 3a) and assembly of polarity complexes
(Fig. 3b) are probably supplemented by an active transport mechanism
that moves proteins from the former uropod into the IS (Fig. 3c).
Surface movements exceeding the rate of diffusion have been noted for
TCRs and for beads moving along the cell surface 61 and have been shown
to depend on phosphatidylinositol-3-OH kinase and myosin 3 . Conversely,
movement of SLP-76 into the central IS seems to rely on microtubules 62 ,
suggesting that a microtubule-based movement, perhaps mediated by a
kinesin, also functions to reorient critical signaling proteins.
Stabilization of polarity in the IS
In a stable IS, actin and actomyosin contraction is diverted away from
rapid motility and functions in maintaining the contact. In that configuration,
the bulk of TCRs and many other receptors are polarized toward the
IS 3,10,61,63,64 . However, a second pool of those receptors often accumulates
at the opposite end of the cell, in what is referred to as the ‘distal pole
complex’ 65 . Once established, that asymmetric distribution is apparently
stable for many hours.
Despite being relatively nonmotile, engaged lymphocytes continuously
expand and contract away from the c-SMAC in the peripheral SMAC
(p-SMAC) and distal SMAC (d-SMAC) regions, where new peptide–major
histocompatibility complex ligands might be encountered and where
microclusters of TCRs continue to form 51,66 . One likely model proposes
that actin fibers initiating at or beyond microclusters in the p-SMAC elongate
into the c-SMAC, perhaps as a result of WASP activation ‘downstream’
of TCR signals 67 (Fig. 4a). Engaged integrins in the p-SMAC might also
promote activation of Cdc42 and WASP, as well as of aPKC, as they do
in migrating astrocytes 68 . The elongating actin fibers might encounter
resistance both from elements in the p-SMAC (generating a retrograde
direction of movement that ultimately returns to the c-SMAC) and from
pSMAC-localized myosin motors (which ‘tug’ on fibers, pulling them backward
into the c-SMAC). A selective ‘clutch’ mechanism, perhaps provided
by CD2-associated protein, which has been associated with TCR internalization
69 , may allow microclusters to attach to this moving ‘radial fiber’ and
become localized in the central IS.
Integrins and the microtubule cytoskeleton are also repolarized after
TCR engagement, and the MTOC faces the IS (Fig. 4b,c). The general
mechanisms underlying MTOC movement are complicated and probably
vary among cell types. For instance, in 3T3 fibroblasts initiating movement
into a wound, the MTOC actually remains stationary while the nucleus is
pushed backward by retrograde actin flow, thereby placing the MTOC in
front of the nucleus. PAR-6 and aPKC are needed to hold the MTOC in
position, whereas Cdc42, myotonic dystrophy–related kinase and myosin
function are required for nuclear relocation 70 . Conversely, in astrocytes, the
MTOC is actively pulled forward by microtubules by a mechanism that
depends on Cdc42, PAR-6–aPKC and dynein and that involves capture of
1146 VOLUME 7 NUMBER 11 NOVEMBER 2006 NATURE IMMUNOLOGY
a b c d e
Microtubule cytoskeleton Surface proteins Vesicle traffic
© 2006 Nature Publishing Group http://www.nature.com/natureimmunology
the microtubule ‘plus’ ends at the leading edge of the cell by adenomatosis
polyposis coli protein and the polarity protein Dlg1 (refs. 68,71,72). The
mechanism of MTOC reorientation in T cells remains to be elucidated.
Figure 4 Five polarity systems in established T cell couples. As in Figure 1, polarity can be viewed from the perspective of five different types of events.
(a,b) The hypothesized ‘radial fibers’ of the actin cytoskeleton (a) and tubulin cytoskeleton (b) represent the profound changes in cytoskeletal reorganization
that occur. (c) Integrins and signaling receptors are ultimately arrayed mainly in the IS as overlapping densities. (d) Many types of vesicles are transported
toward the IS, and we suggest the existence of an additional recycling pool at or near the c-SMAC. (e) Polarity proteins are arrayed in patterns that
presumably reinforce IS stability.
Polarized endocytosis and exocytosis in the IS
Controlled internalization and delivery of vesicles between the cytoplasm
and the plasma membrane is a key component regulating the polarity and
function of motile as well as engaged T cells. For example, in cytotoxic
T lymphocytes, secretory vesicles containing cytotoxic granules line up
alongside the TCR in the IS 73 . Secretory vesicles bearing the cytokines
interleukin 2 and interferon-γ also line up facing the APC, although others
(such as those containing tumor necrosis factor) do not demonstrate
obvious directional ‘preferences’ 11 . Secretory vesicles bearing interleukin 2
and interferon-γ localize together with vesicles defined by Rab proteins as
well as by the v-SNARE syntaxin 6 (ref. 11).
Vesicular release is probably also important for the reinforcement or
modulation of signaling at the IS. Studies of Lck 74 , Lat 75 , TCR 10 and
CTLA-4 (ref. 76) has shown that vesicles bearing those components of the
proximal signaling cascade traffic toward the center of the IS, where they
presumably fuse with the plasma membrane and deliver vesicular content.
Cytokine receptors as well as ‘housekeeping’ proteins such as transferrin are
similarly brought to the IS 64,77 . The recycling of TCRs in vesicles may be
essential for TCR accumulation at the IS 10 , although direct movement of
the TCR into the IS along the plasma membrane 3,61 also occurs. SNAP-23
and syntaxin 4, both of which are components of t-SNAREs, are enriched
in plasma membrane regions at or near the TCR accumulation in the IS,
whereas vesicles bearing cellubrevin and v-SNAREs localize on vesicles
inside the cytoplasm, juxtaposed to the IS 10 . In summary, these observations
suggest the existence of an exocytic machine that delivers proteins
into the IS.
Perhaps equally important to exocytosis is the uptake of proteins by
endocytosis in regions adjacent to or in the IS. It is believed that internalization
occurs at the center of the IS to facilitate TCR degradation 69,78 .
Alterations in CD2-associated protein, for example, result in decreased
TCR internalization 69,79 , and this protein has been associated with ubiquitin
E3 ligases such as Cbl. However, more than simple degradation may
be occurring at the IS. Continuous ruffling of membrane in the peripheral
region as well as coalescing flow toward the central zone may require
consistent reintroduction of membrane at the far outer domains. We thus
suggest that a vesicular cycle (Fig. 4d) may serve to recycle proteins through
the IS during serial receptor-ligand engagements.
How does exocytosis coordinate with the other polarity participants
present in the established IS? Lgl associates with syntaxin-4 in mammalian
cells 80 and may therefore influence the placement of t-SNARE on the
plasma membrane in the peripheral IS, into which proteins may be recycled
(Fig. 4e). Lgl protein may associate with cellular myosins, and such an interaction
may influence the placement of adjacent t-SNAREs and thereby also
influence the adhesiveness of integrins. Different t-SNAREs are required
for the sorting of vesicles to distinct regions of the plasma membrane.
For example, syntaxin-4 is involved in apical sorting, whereas syntaxin-3
is necessary for basolateral sorting. Distinct Rab proteins have also been
associated with that. For example, Rab17 seems to be required specifically
for apical sorting 81 . Rab13 has a specialized function in the recycling and
delivery of transmembrane proteins that form the tight junctions of epithelial
cells, such as occludin 82 . In cytotoxic T lymphocytes, the delivery of lytic
granules to the IS depends on Rab27a, which is important for regulated
secretion in several cell types 73 , and IS-localized secretory vesicles containing
interleukin 2 are associated with Rab3d and Rab19 (ref. 11). Although
syntaxin-4 accumulates along the IS near the TCR 10 , the specialized Rab
proteins involved in recycling membranes at the IS are not yet known.
Although re-establishing polarity during signaling remains a focus of ongoing
work in this area, there is yet another arena in which polarity and its
modification is likely to be potentially more clinically relevant. Underlying
normal T cell surveillance is presumably a coordinated series of cues for
the T cell to follow. Although the organization of those cues remains to
be completely elucidated, it is certain that chemokine gradients into and
throughout organs serve to orient T cell polarity (Fig. 1). The presence
and density of integrin substrates provides a potential ‘highway’ on which
T cells might ‘crawl’. The presentation of chemokines together on an
integrin receptor–bearing surface 83 or to glycosaminoglycans on APCs 84
may also influence T cell function. Two-photon imaging studies have suggested
that T cells might travel along high endothelial venules 85 , perhaps
‘entranced’ by the unique combination of immobilized chemokines and a
local high density of adhesion molecules.
A remaining issue is whether ‘immune-privileged’ sites, such as tumors
or chronic inflammatory sites, might provide a defective series of motility
cues that accounts in part for poor T cell surveillance and tolerance to
antigens produced in these locations. Tumors, for example, co-opt both
chemokine and matrix-remodeling mechanisms that influence the chemotactic
and adhesive pathways discussed above. Might tumors create
T cell disorder by overexpressing some cues while minimizing others?
T cells are known to be much less reactive in their uropod 46 , and loss of
polarity, such as by loss of Scrib, is associated with inhibition of signaling
competence 17 . Notably, depletion of Dlg by means of short hairpin
NATURE IMMUNOLOGY VOLUME 7 NUMBER 11 NOVEMBER 2006 1147
© 2006 Nature Publishing Group http://www.nature.com/natureimmunology
RNA results in overactive T cells in culture 47 , whereas for primary T cell
blasts stimulated with APC plus antigen, loss of Dlg attenuates activation 48 .
Perhaps the difference lies in the polarized cues that accompany the signals
received in those disparate stimulation conditions. As future experiments
examine tolerant environments in greater depth, it may be important to
consider the possibilities that a T cell might well fail to become active or
might remain tolerant precisely because of a disorganized array of conflicting
polarity cues that effectively ‘freezes’ the T cell in a state of inactivity.
We thank J. Gilden for critical reading of the manuscript, and C. Lin for assistance
COMPETING INTERESTS STATEMENT
The authors declare that they have no competing financial interests.
Published online at http://www.nature.com/natureimmunology/
Reprints and permissions information is available online at http://npg.nature.com/
1. Gupton, S.L. et al. Cell migration without a lamellipodium: translation of actin dynamics
into cell movement mediated by tropomyosin. J. Cell Biol. 168, 619–631 (2005).
2. Jacobelli, J., Chmura, S.A., Buxton, D.B., Davis, M.M. & Krummel, M.F. A single
class II myosin modulates T cell motility and stopping but not synapse assembly. Nat.
Immunol. 5, 531–538 (2004).
3. Wülfing, C. & Davis, M.M. A receptor/cytoskeletal movement triggered by costimulation
during T cell activation. Science 282, 2266–2269 (1998).
4. Smith, A. et al. A talin-dependent LFA-1 focal zone is formed by rapidly migrating
T lymphocytes. J. Cell Biol. 170, 141–151 (2005).
5. Serrador, J.M. et al. CD43 interacts with moesin and ezrin and regulates its redistribution
to the uropods of T lymphocytes at the cell-cell contacts. Blood 91, 4632–4644
6. del Pozo, M.A. et al. ICAMs redistributed by chemokines to cellular uropods as a
mechanism for recruitment of T lymphocytes. J. Cell Biol. 137, 493–508 (1997).
7. Krummel, M.F., Sjaastad, M.D., Wülfing, C. & Davis, M.M. Differential assembly of
CD3ζ and CD4 during T cell activation. Science 289, 1349–1352 (2000).
8. Tibaldi, E.V., Salgia, R. & Reinherz, E.L. CD2 molecules redistribute to the uropod
during T cell scanning: implications for cellular activation and immune surveillance.
Proc. Natl. Acad. Sci. USA 99, 7582–7587 (2002).
9. Russell, S. & Oliaro, J. Compartmentalization in T-cell signalling: membrane microdomains
and polarity orchestrate signalling and morphology. Immunol. Cell Biol. 84,
10. Das, V. et al. Activation-induced polarized recycling targets T cell antigen receptors to
the immunological synapse; involvement of SNARE complexes. Immunity 20, 577–588
11. Huse, M., Lillemeier, B.F., Kuhns, M.S., Chen, D.S. & Davis, M.M. T cells use two
directionally distinct pathways for cytokine secretion. Nat. Immunol. 7, 247–255
12. Roche, J.P., Packard, M.C., Moeckel-Cole, S. & Budnik, V. Regulation of synaptic
plasticity and synaptic vesicle dynamics by the PDZ protein Scribble. J. Neurosci. 22,
13. Bilder, D., Li, M. & Perrimon, N. Cooperative regulation of cell polarity and growth by
Drosophila tumor suppressors. Science 289, 113–116 (2000).
14. Montcouquiol, M. et al. Identification of Vangl2 and Scrb1 as planar polarity genes in
mammals. Nature 423, 173–177 (2003).
15. Qin, Y., Capaldo, C., Gumbiner, B.M. & Macara, I.G. The mammalian Scribble polarity
protein regulates epithelial cell adhesion and migration through E-cadherin. J. Cell
Biol. 171, 1061–1071 (2005).
16. Lahuna, O. et al. Thyrotropin receptor trafficking relies on the hScrib-betaPIX-GIT1–
ARF6 pathway. EMBO J. 24, 1364–1374 (2005).
17. Ludford-Menting, M.J. et al. A network of PDZ-containing proteins regulates T cell
polarity and morphology during migration and immunological synapse formation.
Immunity 22, 737–748 (2005).
18. Yamanaka, T. et al. Mammalian Lgl forms a protein complex with PAR-6 and aPKC
independently of PAR-3 to regulate epithelial cell polarity. Curr. Biol. 13, 734–743
19. Plant, P.J. et al. A polarity complex of mPar-6 and atypical PKC binds, phosphorylates
and regulates mammalian Lgl. Nat. Cell Biol. 5, 301–308 (2003).
20. Betschinger, J., Mechtler, K. & Knoblich, J.A. The Par complex directs asymmetric
cell division by phosphorylating the cytoskeletal protein Lgl. Nature 422, 326–330
21. del Pozo, M.A., Vicente-Manzanares, M., Tejedor, R., Serrador, J.M. & Sanchez-Madrid, F.
Rho GTPases control migration and polarization of adhesion molecules and cytoskeletal
ERM components in T lymphocytes. Eur. J. Immunol. 29, 3609–3620 (1999).
22. D’Souza-Schorey, C., Boettner, B. & Van Aelst, L. Rac regulates integrin-mediated
spreading and increased adhesion of T lymphocytes. Mol. Cell. Biol. 18, 3936–3946
23. Haddad, E. et al. The interaction between Cdc42 and WASP is required for SDF-1-
induced T-lymphocyte chemotaxis. Blood 97, 33–38 (2001).
24. Snapper, S.B. et al. WASP deficiency leads to global defects of directed leukocyte
migration in vitro and in vivo. J. Leukoc. Biol. 77, 993–998 (2005).
25. Weiner, O.D. et al. Hem-1 complexes are essential for Rac activation, actin polymerization,
and myosin regulation during neutrophil chemotaxis. PLoS Biol. 4, e38 (2006).
26. Nolz, J.C. et al. The WAVE2 complex regulates actin cytoskeletal reorganization
and CRAC-mediated calcium entry during T cell activation. Curr. Biol. 16, 24–34
27. Yan, C. et al. WAVE2 deficiency reveals distinct roles in embryogenesis and Rac-mediated
actin-based motility. EMBO J. 22, 3602–3612 (2003).
28. Laudanna, C., Campbell, J.J. & Butcher, E.C. Role of Rho in chemoattractant-activated
leukocyte adhesion through integrins. Science 271, 981–983 (1996).
29. Constantin, G. et al. Chemokines trigger immediate β 2 integrin affinity and mobility
changes: differential regulation and roles in lymphocyte arrest under flow. Immunity
13, 759–769 (2000).
30. Volinsky, N., Gantman, A. & Yablonski, D.A. Pak- and Pix-dependent branch of the
SDF-1α signalling pathway mediates T cell chemotaxis across restrictive barriers.
Biochem. J. 397, 213–222 (2006).
31. Weiss-Haljiti, C. et al. Involvement of phosphoinositide 3-kinase γ, Rac, and PAK signaling
in chemokine-induced macrophage migration. J. Biol. Chem. 279, 43273–43284
32. Leeuwen, F.N. et al. The guanine nucleotide exchange factor Tiam1 affects neuronal
morphology; opposing roles for the small GTPases Rac and Rho. J. Cell Biol. 139,
33. Lee, J.H. et al. Roles of p-ERM and Rho-ROCK signaling in lymphocyte polarity and
uropod formation. J. Cell Biol. 167, 327–337 (2004).
34. Palacios, F., Price, L., Schweitzer, J., Collard, J.G. & D’Souza-Schorey, C. An essential
role for ARF6-regulated membrane traffic in adherens junction turnover and epithelial
cell migration. EMBO J. 20, 4973–4986 (2001).
35. Audebert, S. et al. Mammalian Scribble forms a tight complex with the betaPIX
exchange factor. Curr. Biol. 14, 987–995 (2004).
36. Mathew, D. et al. Recruitment of scribble to the synaptic scaffolding complex requires
GUK-holder, a novel DLG binding protein. Curr. Biol. 12, 531–539 (2002).
37. Lue, R.A., Brandin, E., Chan, E.P. & Branton, D. Two independent domains of hDlg are
sufficient for subcellular targeting: the PDZ1–2 conformational unit and an alternatively
spliced domain. J. Cell Biol. 135, 1125–1137 (1996).
38. Wang, H.R. et al. Regulation of cell polarity and protrusion formation by targeting RhoA
for degradation. Science 302, 1775–1779 (2003).
39. Joberty, G., Petersen, C., Gao, L. & Macara, I.G. The cell-polarity protein Par6 links
Par3 and atypical protein kinase C to Cdc42. Nat. Cell Biol. 2, 531–539 (2000).
40. Johansson, A., Driessens, M. & Aspenstrom, P. The mammalian homologue of the
Caenorhabditis elegans polarity protein PAR-6 is a binding partner for the Rho GTPases
Cdc42 and Rac1. J. Cell Sci. 113, 3267–3275 (2000).
41. Lin, D. et al. A mammalian PAR-3-PAR-6 complex implicated in Cdc42/Rac1 and aPKC
signalling and cell polarity. Nat. Cell Biol. 2, 540–547 (2000).
42. Zhang, H. & Macara, I.G. The polarity protein PAR-3 and TIAM1 cooperate in dendritic
spine morphogenesis. Nat. Cell Biol. 8, 227–237 (2006).
43. Chen, X. & Macara, I.G. Par-3 controls tight junction assembly through the Rac
exchange factor Tiam1. Nat. Cell Biol. 7, 262–269 (2005).
44. Habets, G.G. et al. Identification of an invasion-inducing gene, Tiam-1, that encodes a
protein with homology to GDP-GTP exchangers for Rho-like proteins. Cell 77, 537–549
45. Dustin, M.L., Bromley, S.K., Kan, Z., Peterson, D.A. & Unanue, E.R. Antigen receptor
engagement delivers a stop signal to migrating T lymphocytes. Proc. Natl. Acad. Sci.
USA 94, 3909–3913 (1997).
46. Negulescu, P.A., Krasieva, T.B., Khan, A., Kerschbaum, H.H. & Cahalan, M.D. Polarity
of T cell shape, motility, and sensitivity to antigen. Immunity 4, 421–430 (1996).
47. Xavier, R. et al. Discs large (Dlg1) complexes in lymphocyte activation. J. Cell Biol.
166, 173–178 (2004).
48. Round, J.L. et al. Dlgh1 coordinates actin polymerization, synaptic T cell receptor and
lipid raft aggregation, and effector function in T cells. J. Exp. Med. 201, 419–430
49. Faure, S. et al. ERM proteins regulate cytoskeleton relaxation promoting T cell-APC
conjugation. Nat. Immunol. 5, 272–279 (2004).
50. Dulyaninova, N.G., Malashkevich, V.N., Almo, S.C. & Bresnick, A.R. Regulation of
myosin-IIA assembly and Mts1 binding by heavy chain phosphorylation. Biochemistry
44, 6867–6876 (2005).
51. Campi, G., Varma, R. & Dustin, M.L. Actin and agonist MHC-peptide complex-dependent
T cell receptor microclusters as scaffolds for signaling. J. Exp. Med. 202, 1031–
52. Barda-Saad, M. et al. Dynamic molecular interactions linking the T cell antigen receptor
to the actin cytoskeleton. Nat. Immunol. 6, 80–89 (2005).
53. Bunnell, S.C., Kapoor, V., Trible, R.P., Zhang, W. & Samelson, L.E. Dynamic actin
polymerization drives T cell receptor-induced spreading: a role for the signal transduction
adaptor LAT. Immunity 14, 315–329 (2001).
54. Bubeck Wardenburg, J. et al. Regulation of PAK activation and the T cell cytoskeleton
by the linker protein SLP-76. Immunity 9, 607–616 (1998).
55. Hanada, T., Lin, L., Chandy, K.G., Oh, S.S. & Chishti, A.H. Human homologue of the
Drosophila discs large tumor suppressor binds to p56lck tyrosine kinase and Shaker
type Kv1.3 potassium channel in T lymphocytes. J. Biol. Chem. 272, 26899–26904
56. Hanada, T., Lin, L., Tibaldi, E.V., Reinherz, E.L. & Chishti, A.H. GAKIN, a novel kinesinlike
protein associates with the human homologue of the Drosophila discs large tumor
suppressor in T lymphocytes. J. Biol. Chem. 275, 28774–28784 (2000).
1148 VOLUME 7 NUMBER 11 NOVEMBER 2006 NATURE IMMUNOLOGY
© 2006 Nature Publishing Group http://www.nature.com/natureimmunology
57. Lee, K.H. et al. T cell receptor signaling precedes immunological synapse formation.
Science 295, 1539–1542 (2002).
58. Phee, H., Abraham, R.T. & Weiss, A. Dynamic recruitment of PAK1 to the immunological
synapse is mediated by PIX independently of SLP-76 and Vav1. Nat. Immunol. 6,
59. Yablonski, D., Kane, L.P., Qian, D. & Weiss, A.A. Nck-Pak1 signaling module is required
for T-cell receptor-mediated activation of NFAT, but not of JNK. EMBO J. 17, 5647–
60. Macara, I.G. Parsing the polarity code. Nat. Rev. Mol. Cell Biol. 5, 220–231 (2004).
61. Moss, W.C., Irvine, D.J., Davis, M.M. & Krummel, M.F. Quantifying signaling-induced
reorientation of TCRs during immunological synapse formation. Proc. Natl. Acad. Sci.
USA 99, 15024–15029 (2002).
62. Bunnell, S.C. et al. T cell receptor ligation induces the formation of dynamically regulated
signaling assemblies. J. Cell Biol. 158, 1263–1275 (2002).
63. Wülfing, C., Sjaastad, M.D. & Davis, M.M. Visualizing the dynamics of T cell activation:
Intracellular adhesion molecule 1 migrates rapidly to the T cell/B cell interface and
acts to sustain calcium levels. Proc. Natl. Acad. Sci. USA 95, 6302–6307 (1998).
64. Batista, A., Millan, J., Mittelbrunn, M., Sanchez-Madrid, F. & Alonso, M.A. Recruitment
of transferrin receptor to immunological synapse in response to TCR engagement.
J. Immunol. 172, 6709–6714 (2004).
65. Cullinan, P., Sperling, A.I. & Burkhardt, J.K. The distal pole complex: a novel membrane
domain distal to the immunological synapse. Immunol. Rev. 189, 111–122 (2002).
66. Yokosuka, T. et al. Newly generated T cell receptor microclusters initiate and sustain
T cell activation by recruitment of Zap70 and SLP-76. Nat. Immunol. 6, 117–127
67. Cannon, J.L. et al. Wasp recruitment to the T cell:APC contact site occurs independently
of cdc42 activation. Immunity 15, 249–259 (2001).
68. Etienne-Manneville, S. & Hall, A. Integrin-mediated activation of Cdc42 controls cell
polarity in migrating astrocytes through PKCζ. Cell 106, 489–498 (2001).
69. Lee, K.H. et al. The immunological synapse balances T cell receptor signaling and
degradation. Science 302, 1218–1222 (2003).
70. Gomes, E.R., Jani, S. & Gundersen, G.G. Nuclear movement regulated by Cdc42,
MRCK, myosin, and actin flow establishes MTOC polarization in migrating Cells. Cell
121, 451–463 (2005).
71. Etienne-Manneville, S. & Hall, A. Cdc42 regulates GSK-3β and adenomatous polyposis
coli to control cell polarity. Nature 421, 753–756 (2003).
72. Etienne-Manneville, S., Manneville, J.B., Nicholls, S., Ferenczi, M.A. & Hall, A. Cdc42
and Par6-PKCζ regulate the spatially localized association of Dlg1 and APC to control
cell polarization. J. Cell Biol. 170, 895–901 (2005).
73. Stinchcombe, J.C., Bossi, G., Booth, S. & Griffiths, G.M. The immunological synapse
of CTL contains a secretory domain and membrane bridges. Immunity 15, 751–761
74. Ehrlich, L.I., Ebert, P.J., Krummel, M.F., Weiss, A. & Davis, M.M. Dynamics of p56lck
translocation to the T cell immunological synapse following agonist and antagonist
stimulation. Immunity 17, 809–822 (2002).
75. Bonello, G. et al. Dynamic recruitment of the adaptor protein LAT: LAT exists in two distinct
intracellular pools and controls its own recruitment. J. Cell Sci. 117, 1009–1016
76. Egen, J.G. & Allison, J.P. Cytotoxic T lymphocyte associated antigen (CTLA-4) accumulation
in the immunological synapse is regulated by TCR signal strength. Immunity
16, 23–35 (2002).
77. Maldonado, R.A., Irvine, D.J., Schreiber, R. & Glimcher, L.H. A role for the immunological
synapse in lineage commitment of CD4 lymphocytes. Nature 431, 527–532
78. Varma, R., Campi, G., Yokosuka, T., Saito, T. & Dustin, M.L. T cell receptor-proximal
signals are sustained in peripheral microclusters and terminated in the central supramolecular
activation cluster. Immunity 25, 117–127 (2006).
79. Kirsch, K.H. et al. The adapter type protein CMS/CD2AP binds to the proto-oncogenic
protein c-Cbl through a tyrosine phosphorylation-regulated Src homology 3 domain
interaction. J. Biol. Chem. 276, 4957–4963 (2001).
80. Musch, A. et al. Mammalian homolog of Drosophila tumor suppressor lethal (2)
giant larvae interacts with basolateral exocytic machinery in Madin-Darby canine
kidney cells. Mol. Biol. Cell 13, 158–168 (2002).
81. Zacchi, P. et al. Rab17 regulates membrane trafficking through apical recycling
endosomes in polarized epithelial cells. J. Cell Biol. 140, 1039–1053 (1998).
82. Terai, T., Nishimura, N., Kanda, I., Yasui, N. & Sasaki, T. JRAB/MICAL-L2 is a
junctional Rab13-binding protein mediating the endocytic recycling of occludin.
Mol. Biol. Cell 17, 2465–2475 (2006).
83. Shamri, R. et al. Lymphocyte arrest requires instantaneous induction of an
extended LFA-1 conformation mediated by endothelium-bound chemokines. Nat.
Immunol. 6, 497–506 (2005).
84. Friedman, R.S., Jacobelli, J. & Krummel, M.F. Surface-bound chemokines capture
and prime T cells for synapse formation. Nat. Immunol. 7, 1101–1108 (2006).
85. Mempel, T.R., Henrickson, S.E. & von Andrian, U.H. T-cell priming by dendritic
cells in lymph nodes occurs in three distinct phases. Nature 427, 154–159
NATURE IMMUNOLOGY VOLUME 7 NUMBER 11 NOVEMBER 2006 1149